Claude Opus 5 (xHigh): benchmark results
Provider: Anthropic. Released 2026-07-24. Access: API.
Unified ELO 1791 ± 1, rank #10 of 1761 rated models, from 14 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ProgramBench | 4.5 | Resolved (%) | 100 |
| ProgramBench Almost | 37 | Almost (%) | 100 |
| Creative Writing (Lechmazur) | 4.1 | Mean Score | 98.9 |
| DuelLab Overall | 76 | DuelLab Score | 98.2 |
| Design Arena (Agents - Mobile Apps) | 1298 | Elo | 97.7 |
| MCP Atlas | 85.8 | Pass Rate (self-reported) | 97 |
| LLM2014 Logic 2026-09 | 64.42 | Median Score | 95.5 |
| SEAL - MCP Atlas | 85.8 | Score | 93.3 |
| LLM2014 Logic 2026-08 | 64.74 | Median Score | 93.2 |
| LLM2014 Logic 2026-07 | 71.88 | Median Score | 93 |
| Epoch AI - Critpt | 27.71 | Score | 92.6 |
| CursorBench 3.1 | 69.3 | Score (%) | 90.4 |
Interactive version: theaggregate.ai/model?slug=claude-opus-5-xhigh · How It Works · Data refreshed daily, snapshot 2026-09-05.